Costs and results of cardiac operations in infants less than 4 months old

J

THoRAc CARDIOVASC SURG

91:667-673, 1986

Costs and results of cardiac operations in infants less than 4 months old Are they worthwhile? From 1979 through 1983,328 of 1,388 pediatric cardiac operations involved patients undergoing their first procedure at less than 4 months of age. Of these, 220 patients had 265 nonductal procedures, and their case historiesare reviewed for results and total hospitalcost. Initial operative mortality was 20 % (43 patients~ Infants with lower operative age and operative weight tended to have closed procedures. Mortality and cure were not related to gestational age, birth weight, age at operation, number of operations, or type of operation. Lower operative weight was associated with a greater mortality. Evaluated survivors (142 patients) were foUowed for a mean of 24 months. Fifteen percent (33 patients) died during foUow-up. Of survivors, 80% (114 patients) had optimized general health; a subset of 29% had normal cardiac function, and 17 % were cured. Lower birth weight was associated with curable lesions and normalcy (p < 0.04). Longer preoperative hospital stay and lower weight at operation were associated with higher hospital cost (p < 0.05). Hospital cost was not related to type of operation, gestational age, birth weight, age at operation, mortality, cure, or normalcy. Acquired neurologic dysfunction and long-term disability were uncommon. The mean hospital cost for surviving infants was $80,000 (1984 doUars). Effective hospital cost per survivor was $110,000. Mortality, cure, and normal function after cardiac operations in infants less than 4 monthsof age werenot related to gestational age, birth weight, or age at operation. Mortality was higher in patients with a lower weight at operation. Separation into distinct fiscal cost groups is not reasonable in this series. Because most survivors are in normal or optimized cardiac health, intensive cardiovascular care in this population is justified.

Donald C. Watson, Jr., M.D., Memphis, Tenn., Linda M. Bradley, M.D., Frank M. Midgley, M.D., and Lewis P. Scott, M.D., Washington, D. C.

h e operative management of neonates and young infants with cardiac disease has been fully described.':' Long-termresults of treatment in these patients has also been documented."!' Recently, however, the economic impact and effectiveness of cardiovascular operations in patients with arteriosclerotic cardiovascular disease has been the topic of considerable debate. I 2·20 In the current milieu of fiscal responsibility":" and optimization of

From LeBonheur Children's Medical Center. University of Tennessee Center for Health Sciences, Memphis, Tenn., and Children's Hospital National Medical Center, George Washington University School of Medicine, Washington, D. C. Read at the Eleventh Annual Meeting of The Western Thoracic Surgical Association, Incline Village, Nev., June 16-20, 1985. Address for reprints: Donald C. Watson, Jr., M.D., Chairman, Cardiothoracic Surgery, 956 Court Ave., Suite 230, Memphis, Tenn. 38163.

medical care, particularly in neonatal patients,":" we investigated the mortality, morbidity, cost, and medical results of care for infants who received a nonductal cardiac operation at less than 4 months of age at the Children's Hospital National Medical Center. Methods From January, 1979, through December, 1983, a cohort of 220 infants who underwent their first nonductal cardiac operation at less than 4 months of age had 265 operations on the heart and great vessels (Fig. 1). These proceduresconstituted 19%of all cardiac surgical operations during the study interval. Of these operations, 229 were performed when the infant was less than 4 months of age and form the basis of this study. Operative mortality was definedas death within 30 days of operation and late mortality as death more than 30 days after operation. The outcome of survivors was 667

The Journal of

668

Watson et af.

Thoracic and Cardiovascular Surgery

Table II. Analyzed variables by operative mortality group in all infants less than 4 months old undergoing nonductal cardiac operations

220 PATIENTS 265 OPERATIONS

Nonsurvivors (N=43)

Variable· Gestational age (wk) Birth weight (kg) Operative age (days) Operative weight (kg) Preoperative stay (day) Cost (1984 dollars)

39 3.2 40 3.6 6.9 80,000

38 3.0 32 3.3 5.2 57,000

p Value

NS NS NS 0.01

NS NS

Legend: NS means p > 0.05.

Fig. 1. Study group outlined. Note that nine patients in the study group had both an open and closed procedure. Open means cardiopulmonary bypass was used. Closed means without use of cardiopulmonary bypass.

Table I. Variable profile for all infants undergoing nonductal cardiac operations at less than 4 months of age

n

Variable Gestational age (wk) Birth weight (kg) Operative age (days) Operative weight (kg) Preoperative stay (days) Cost (1984 dollars) Follow-up interval (wk)

Value 39 3.1 37 3.5 6.3 73,000 68

39 3.2 58 3.9 11 66,000 64

NS NS 0.001 0.001 0.001

NS NS

General health was categorized as normal, optimized, or impaired. Patients in optimized general health were cured or had normal and effectively palliated cardiovascular status and no evidence of obvious neurologic impairment. Impaired health indicated either major acquired handicaps not present preoperatively or congenital disabilities. Total hospital cost was determined from review of inpatient hospital and physician billings during the 5 year study interval. Cost was corrected for inflation and expressed in thousands of 1984 dollars. 28- 32 Statistical analysis was performed by the Student's t test for independent means, Fisher's exact test, correlation, and regression analysis. A probability value of less than 0.05 was considered significant.

Legend: NS means p > 0.05.

Results

'Nine infants underwent both an open and closedprocedure in the first 4 months of life.

The patient profile is described in Table I. Operative age and weight were greater and preoperative hospital stay was longer in infants having open operations compared with those having closed procedures. Of course, operative weight was strongly correlated with operative age (r = 0.6, p < 0.001). There were 43 operative deaths, 18 in the group with open procedures and 25 in those with closed procedures, for an overall operative mortality of 20%. The relationship of preoperative variables to mortality is depicted in Table II. The only analyzed variable associated with increased operative mortality was lower operative weight. In the subgroup of infants undergoing bypass procedures (Table III), lower operative age also was associated with a significantly increased risk of death. This finding was not present in the subgroup of patients having closed operations. A total of 97 cardiopulmonary bypass procedures were performed on a subgroup of 77 patients. Thus, 20 nonprimary open procedures were performed on this patient subgroup during the 5 year interval. All opera-

determined by retrospective analysis of their medical records and correspondence with parents or physicians, or both. Follow-up time was the interval from operation to the last evaluation before April 15, 1984. Diagnoses were established by cardiac catheterization, at operation, or at autopsy in all patients. Gestational age, birth weight, operative age, operative weight, number of preoperative hospital days, and operation type were analyzed for their relationship to mortality, morbidity, and total hospital cost (Fig. 2). Survivors were assessed for cardiovascular status and overall general health. Cardiovascular status was defined as normal or abnormal on the basis of the presence or absence of structural or electrophysiologic residua. Patients were considered to have optimized cardiovascular status if they had received effective palliation or treatment of their congenital heart disease.

Volume 91 Number 5 May, 1986

Cardiac operations in infants

Table m. Analyzed variables by operative mortality group in infants less than 4 months old undergoing cardiopulmonary bypass procedures Variable Gestational age (wk) Birth weight (kg) Opera tive age (days) Operative weight (kg) Preoperative stay (day) Cost (1984 dollars)

Nonsurvivors IN = 18)

p Value

38.5 3.1 37 3.5 6.1 55,000

NS NS 0.01 O.oI NS NS

Legend: NS means p> 0.05.

tions were done with surface and core cooling and rare use of deep hypothermia with total circulatory arrest. Table IV shows the lesions treated and mortalities at the first open operation. The overall operative mortality at the first operation in this group was 23%. An additional 9% died during the follow-up interval. A total of 168 nonbypass procedures were performed on a subgroup of 152 patients. Thus, 16 nonprimary closed procedures were performed on this patient subgroup. Nine patients had both an open and closed procedure in the first 4 months of life. Fifteen infants underwent both coarctation repair and another closed procedure (i.e., pulmonary artery band or aorto-pulmonary shunt) and were classified in the group with coarctations. One infant underwent vascular ring division and shunt placement and was classified in the group with decreased pulmonary flow. Table V describes the lesions treated and mortalities at the first operation for infants having nonbypass procedures. Coarctation complex included isolated aortic coarctation, coarctation with ventricular septal defect, and coarctation with complex intracardiac anatomy. Lesions with decreased pulmonary blood flow were tetralogy of Fallot, tricuspid atresia, single ventricle with pulmonary stenosis, doubleoutlet right ventricle with pulmonary stenosis, Ebstein's anomaly, and atrioventricular canal with pulmonic stenosis. Anomalies with excessive pulmonary blood flow included single ventricle, ventricular septal defect, truncus arteriosus, tricuspid atresia, atrioventricular canal, and dextro-transposition of the great vessels with ventricular septal defect. Other patients having nonbypass procedures were infants with complete heart block, absent right pulmonary artery, aortic stenosis, and left ventricular rhabdomyoma. There were 25 operative deaths at the first closed operation, a mortality of 16%. An additional 25 patients (16%) died during followup. The evaluation of cardiovascular status for 142

669

PREOP

.1

-,

GAGE

f-----.

PREOPD

OUTCOME

MORT B

WGT

::OPTYPE~

CURE

OP AGE NORM OP WGT

.1 '1

~

COST

Fig. 2. Relationships examined: B WGT, Birth weight. COST, Total hospital cost during study. CURE, Cure rate during follow-up. G AGE, Gestational age. MORT, Mortality rate during follow-up (early and late). NORM, Normalcy rate during follow-up. OP AGE, Age at operation. OP WGT, Weight at operation. OPTYPE, Type of operation (open versus closed). PREOPD, In-hospital days before operation.

Table IV. Mortality rates by lesion in infants less than 4 months old undergoing cardiopulmonary bypass procedures

Lesion

No. of operations

VSD AVC PS TAPVD d-TGA Other*

17

Total

77

13 13 12 9 13

Operative deaths No.

I

Late deaths

%

No. 0

7

12 23 15 33 0 54

1 1 1 3

0 8 8 8 11 23

18

23

7

9

2 3 2 4 0

J

I %

Legend: VSD, Ventricular septal defect. AVe, Atrioventricular canal. PS, Pulmonic stenosis. TAPVD, Total anomalous pulmonary venous drainage. d-TGA, Dextro-transposition of the great arteries. 'Other: Anomalous left coronary artery (2); aortic stenosis (l); double-outlet right ventricle '(2); hypoplastic left heart syndrome (2); hypoplastic right heart syndrome (J); levo-transposition of the great arteries (J); pulmonary artery sling (J); truncus arteriosus (2); tetralogy of Fallot with absent pulmonary valve (I).

long-term survivors was complete. Status was normal in 41 patients (29%) and abnormal in 101 patients (71%). The normal group included 13 patients with isolated coarctation of the aorta, 11 with ventricular septal defect, six with total anomalous pulmonary venous drainage, six with vascular ring, three with atrioventricular canal, and two patients with pulmonic stenosis with pressure gradients less than 15 torr. Normal cardiac outcome correlated only with lower birth weight and

The Journal of

670

Watson et al.

Thoracic and Cardiovascular Surgery

Table V. Mortality rates by lesion in infants less than 4 months old undergoing closed nonductal cardiac procedures

Abnormality Coarctation complex Decreased pulmonary flow Increased pulmonary flow Vascular ring Other* Total

No. of operations

Operative deaths

Late deaths

No·1 %

No·1 %

58

II 10

18 7 6

2 I I

17 17 II 14 17

152

25

16

63

6 II 6 0 2

10 19 33 0 33

25

16

'Other: Complete heart block (3); absent right pulmonary artery (I); aortic stenosis (I); left ventricular rhabdomyoma (I).

Table VI. Analyzed variables by cardiovascular

Discussion

status for long-term survivors after nonductal cardiac operations at less than 4 months of age p Variable Gestational age (wk) Birth weight (kg) Opera tive age (days) Operative weight (kg) Preoperative stay (days) Cost (1984 dollars)

tion (80%), and a subset of 32 infants were in normal condition (23%). Twenty-four survivors had congenital impairments (17%). The cardiovascular status in these patients had been optimized. Only four infants had acquired neurologic handicaps (3%). Total hospital cost was compared between open and closed procedures (Table I), surviving and nonsurviving infants (Tables II and III), and patients with normal versus abnormal cardiovascular status (Table VI). No significant differences were found. The only preoperative variable that correlated with higher total hospital cost was length of preoperative hospitalization. The mean total hospital cost for long-term survivors was $80,000. When the cost of the nonsurvivors was distributed over the surviving group, the effective total hospital cost per survivor was $110,000.

Value

NS 0.04

NS NS 0.02

NS

Legend: NS means p> 0.05.

longer preoperative stay (Table VI). Mean total hospital cost was almost 50% higher in patients with abnormal cardiovascular status. The large absolute difference in cost was not statistically significant when patients with normal cardiovascular status were compared with those with abnormal status, because of the large standard deviation in the abnormal group; three surviving patients with cardiac abnormalities had costs exceeding $400,000. Thirty-two of 101 infants classified as having abnormal cardiovascular status had only mild residual hemodynamic or electrophysiologic disturbances. Included were patients with mild pulmonic insufficiency (13), small residual ventricular septal defect (six), mild superior vena caval obstruction (five), elevated pulmonary vascular resistance (four), complete heart block (two), sinus node dysfunction (one), and mild aortic stenosis (one). The remaining 69 survivors had significant residual disease. Overall general health of the survivors was reviewed. One hundred fourteen infants were in optimized condi-

We investigated the cost and effects of 229 nonductal operations on 220 patients less than 4 months of age with cogenital heart lesions. Seventy-seven infants underwent operations performed with cardiopulmonary bypass, and there were 18 (23%) operative deaths. Four deaths were due to uncorrectable anatomy (hypoplastic left heart in two and truncus arteriosus with truncal valve stenosis in two), three occurred in patients moribund preoperatively, and one was caused by liver failure resulting from unrecognized biliary atresia. These deaths appear unrelated to the use of cardiopulmonary bypass. Low cardiac output precipitated the remaining 10 operative deaths. The adjusted operative mortality was 13%. One hundred fifty-two infants underwent nonbypass procedures, with 25 (16%) deaths. Four were due to uncorrectable anatomy (hypoplastic left heart variants in three and diffuse ventricular tumor in one), two were due to catheterization complications, and two occurred in patients moribund preoperatively. The adjusted operative mortality was 11%. If cardiopulmonary bypass has a selective damaging effect on the very young,' then adjusted operative mortality ought to be significantly higher in the cardiopulmonary bypass group. We did not find a detrimental effect of cardiopulmonary bypass. In this series operative mortality appears to be lesion specific. Given the comparable mortalities for open and closed procedures, our experience suggests that the trend toward early total surgical correction is justified. 1,2, 8, 33 Analysis of preoperative variables for the whole group reveals only lower operative weight to be associated with increased risk of death. When the bypass group alone is considered, however, lower operative age yields a higher

Volume 91 Number 5 May, 1986

risk of mortality. This suggests that the very young who require early bypass operations may have more devastating disease. The 142 survivors with completed evaluation were followed for a mean of 24 months. The cardiovascular status ofsurvivorsundergoing open procedures was normal in 22 (42% of 52 open survivors) and markedly improved in all. Only four of the infants having open operations have required reoperation. The cardiac status of surviving infants who did not undergo bypass was frequently abnormal, because most of these patients had initial surgical palliation for complex lesions or were not consideredgood candidates for total surgical correction. The overall general health of survivors was remarkably good. Eighty percent of the surviving infants were either in normal or in optimized general health. The cardiovascular status in survivors had been optimized in 97% of patients. The majority of patients whose condition was optimized were infants with palliated complex disease. Seventeen percent of survivors, however, had congenital impairments that adversely affected their general health. Only 3% of surviving infants had clear neurologic injury at operation. Because of the relatively short follow-up interval in this series, more subtle neurologic abnormalities'" were not examined. One important question not addressed in this study is whether the cardiovascular morbidity for infants undergoing operation at less than 4 months is greater than it is for older infants. Other studies show a reduced risk of mortality and morbidity with advancing age for most nonlethal congenital abnormalities. 2, 4, 9-11 When the costs for treatment in this cohort are compared to costs for other critically ill neonates," a more optimistic situation can be seen. This is accentuated by the belief that many of the critically ill infants with cardiac disease can be expected to have a normal life expectancy and productivity after curative treatment. The cost-benefit ratio for these infants should be considered worthwhile. Another issue not discussed herein is a concern about the widening gap between billed medical services and payment for these services. In the next decade, we will have to deal with many complex ethical, moral, legal, and fiscal problems. Health care should be optimized for all patients, but costs may restrict our ability to do everything for everyone at all times. A clear definition of certain high-cost low-yield medical problems may help in the allocation of national resources. We thank Grant Somes, Ph.D., Associate Professor of Biostatistics, University of Tennessee, for his thorough review and analytic recommendations.

Cardiac operations in infants 6 7 1

REFERENCES Bove EL, Behrendt DM: Open-heart surgery in the first week of life. Ann Thorac Surg 29: 130-134, 1979 2 Bove EL, Bull C, Stark J, de Leval M, Macartney FJ, Taylor JFN: Congenital heart disease in the neonate. Results of surgical treatment. Arch Dis Child 58: 137-141, 1983 3 Jonas RA, Castanada AR, Freed MD: Normothermic caval inflow occlusion. Application to operations for congenital heart disease. J THORAC CARDIOVASC SURG 89:780-786, 1985 4 Kirklin JK, Blackstone EH, Kirklin JW, McKay R, Pacifico AD, Bargeron LM: Intracardiac surgery in infants under age 3 months. Predictors of postoperative in-hospital cardiac death. Am J Cardiol 48:507-512, 1981 5 Naito Y, Fugita T, Tomino T, Yoshiharu K, Isobe F, Hayashi K, Kito Y, Manabe H, Kamiya T: Surgical treatment for severe congenital heart diseases. Jpn Circ J 47:1137-1146, 1983 6 Stark J, Smal1horn J, Huhta J, de Leval M, Macartney FJ, Rees PG, Taylor JFN: Surgery for congenital heart defects diagnosed with cross-sectional echocardiography. Circulation 68:Suppl 2:129-138, 1983 7 Sugimura S, Starr A: Cardiopulmonary bypass in infants under four months of age. J THoRAc CARDIOVASC SURG 73:894-899, 1977 8 Turley K, Mavroudis C, Ebert PA: Repair of congenital cardiac lesions during the first week of life. Circulation 66:Suppl 1:214-219, 1982 9 Dickinson DF, Arnold R, Wilkinson JL: Outcome of treatment for neonates referred to a supraregional cardiac centre 1976-78. Arch Dis Child 57:328-333, 1982 10 Fyler DC, ed.: Report of the New England Regional Infant Cardiac Program. Pediatrics 65:Suppl 375-461, 1980 1I Macartney FJ, Taylor JFN, Graham GR, de Leval M, Stark J: The fate of survivors of cardiac surgery in infancy. Circulation 62:80-91, 1980 12 Arnold CB, ed.: Costs of coronary bypass surgery. Stat Bull 66:2-9, April-June, 1985 13 Cohen HA, Solnick M, Stephenson A: The financing of coronary artery bypass surgery. Circulation 66:Suppl 3:49-55, 1982 14 Collins JJ, Tucker WY, Kopf G, Koster JK, Mee RBB, Cohen LH: The impact of revascularization surgery upon hospital costs in patients with angina pectoris. Am J Cardiol 41:447, 1978 15 Just-Viera JO, Bunker FH: Cardiac surgery in a small community. An eight-year experience. Ann Thorac Surg 33:212-217,1982 16 Kronenfeld JJ, Charles ED, Wayne lB, Oberman A, Kouchoukos NT, Rogers WJ, Mantle JA, Rackley CE, Russel1 RO: Unstable angina pectoris. An examination of modes and costs of therapy. Circulation 60:Suppl 1:16-22, 1979

The Journalof Thoracic and Cardiovascular Surgery

6 7 2 Watson et al.

17 Loop FD, Christiansen EK, Lester JL, Cosgrove DM, Franco I, Golding LR: A strategy for cost containment in coronary surgery. JAMA 250:63-66, 1983 18 Roberts AJ, Woodhall DD, Conti CR, Ellison DW, Fisher R, Richards C, Marks RG, Knauf DG, Alexander JA: Mortality, morbidity, and cost-accounting related to coronary artery bypass graft surgery in the elderly. Ann Thorac Surg 39:426-432, 1985 19 Stoney WS, Alford WC, Burrus GR, First RA, Thomas CS: The cost of coronary bypass procedures. JAMA 240:2278-2280, 1978 20 Weinstein MC, Stason WB: Cost-effectiveness of coronary artery bypass surgery. Circulation 66:Suppl 3:56-65, 1982 21 Bloom BS, Knorr RS, Evans AE: The epidemiology of disease expenses. JAMA 253:2393-2397, 1985 22 Drucker WR, Gavett JW, Kirshner R, Messick WJ, Ingersoll G: Toward strategies for cost containment in surgical patients. Ann Surg 198:284-300, 1983 23 Zook CJ, Moore FD: High-cost users of medical care. N Engl J Med 302:996-1002, 1980 24 Madoff RD, Sharpe SM, Fath JJ, Simmons RL, Cerra FB: Prolonged surgical intensive care. Arch Surg 120:698702, 1985 25 Dawodu AH, Effiong CE: Neonatal mortality. Effects of selective pediatric interventions. Pediatrics 75:51-57, 1985 26 McCormick MC: The contribution of low birth weight to infant mortality and childhood morbidity. N Engl J Med 312:82-90, 1985 27 Walker DB, Feldman A, Vohr BR, Oh W: Cost-benefit analysis of neonatal intensive care for infants weighing less than 1,000 grams at birth. Pediatrics 74:20-25, 1984 28 U.S. Department of Labor, Bureau of Labor Statistics: Consumer Price Index. All Urban Consumers (CPI-U), U.S. City Average Medical Care (1967=100). Washington, D. C., U.S. Bureau of Labor Statistics, December, 1983 29 U.S. Department of Labor, Bureau of Labor Statistics: Consumer Price Index. All Urban Consumers (CPI-U), U.S. City Average Medical Care (1967=100). Washington, D. c., U.S. Bureau of Labor Statistics, December, 1982 30 U.S. Department of Labor, Bureau of Labor Statistics: Consumer Price Index. All Urban Consumers (CPI-U), U.S. City Average Medical Care (1967=100). Washington, D. C, U.S. Bureau of Labor Statistics, December, 1981 31 U.S. Department of Labor, Bureau of Labor Statistics: Consumer Price Index. All Urban Consumers (CPI-U), U.S. City Average Medical Care (1967=100). Washington, D. C., U.S. Bureau of Labor Statistics, December, 1980 32 U.S. Department of Labor, Bureau of Labor Statistics: Consumer Price Index. All Urban Consumers (CPI-U), U.S. City Average Medical Care (1967=100). Washing-

ton, D. c., U.S. Bureau of Labor Statistics, December, 1979 33 Sade RM, Crawford FA, Hohn AR: Inflow occlusion for semilunar valve stenosis. Ann Thorac Surg 33:570-575, 1982 34 Newburger JW, Silbert AR, Buckley LP, Fyler DC: Cognitive function and age at repair of transpostion of the great arteries in children. N Engl J Med 310:1495-1499, 1984

Discussion DR. D. CRAIG MILLER Stanford. Calif

Dr. Watson has touched on some important issues, many of which are societal and philosophical. For example, in China it is tacitly understood that neonates will not consume national resources for correction of congenital heart disease. I am told this is also true in Russia. This sort of "natural selection" leaves the surgeons in these countries with beautiful patient populations; for example, hundreds of low-risk, naturally selected patients with tetralogy of Fallot are on waiting listsin China. In more developed parts of the world, however, it has not been judged to be reasonable or justifiable to adopt this policy,at least so far, but Dr. Watson's presentation does make us address these issues. Some of my more cynical colleagues have suggested that the whole field of neonatal congenital heart disease was born only after the large numbers of older patients with congenital heart disease declined. I do not think that is altogether true. Quite seriously, there have been great advances due to open heart surgery in neonates, particularly in patients with atrioventricular septal defects, truncus arteriosus, and total anomalous pulmonary venous return. Perhaps even more important is the issue of increasing pressure to limit medical costs. We as physicians may soon be asked not only to make decisions regarding medical risk, but also to decide whether or not we and the hospital can afford the financial risk of attempting to treat some of these newborn babies. Dr. Watson, I have three questions. I am confused about the skewed nature of the patient population. You operated on 13 patients with atrioventricular septal defects within the first 4 months of life, which parallels our experience at Stanford. However, in other (higher elevation) cities, surgeons do not appear to see these patients in infancy. On the other hand, I did not see many cases of truncus arteriosus in your series,and I could not discern how many patients with hypoplastic left heart syndrome you had attempted to treat. Second, on the basis of your study, have you and your colleagues in Washington, D.C. (or now in you new practice in Memphis) been able to change patient treatment or management in terms of both optimizing medical risk and minimizing costs? For example, would it not be more cost effective in a 2-day-old neonate with simple transposition to perform a switch operation and avoid balloon atrial septostomy and a subsequent intra-atrial (Senning or Mustard) repair? It makes

Volume 91 Number 5 May, 1986

sense costwise, may well make sense physiologically, but do we really know? Third, as you have intimated, follow-up is short in your paper. Can you really use the word "cure?" I am always cautious when I see that, but it is important here, because exactly how many of these patients are really going to go on and be tax payers over the next 50 to 60 years and repay this debt to society? Assuming the result is as optimal as you extrapolate, a $100,000 investment amortized over 60 or 70 years is cheap, and clearly much more rational and reasonable than a $100,000 investment late in life. Finally, could you speculate concerning what should be done about the hidden, indirect subsidization of the treatment of congenital heart disease in terms of the broad gap between hospital cost and hospital revenue? DR. RICHARD M. PETERS Sail Diego, Calif

Dr. Watson, in closing would you discuss the long-term projected outcome for those patients who had the highest cost of $400,000 and give more details about those three patients? DR. WATSON (Closing) Thank you, Dr. Miller, for your review of the paper and your discussion of some of the alternatives. Truncus arteriosus was handled by a closed technique in this series. That philosophy has changed, though, and we probably handle this anomaly differently now. I believe the operation to correct hypoplastic left ventricle currently is an experimental procedure. A lot of people disagree. I believe the anomaly should be handled at major centers, maybe three or four throughout the country, and until

Cardiac operations in infants 6 7 3

the major pitalls of that particular procedure are defined and solved, others of us should not undertake that correction. We had a brief symposium right before the AATS meeting, and the overall mortality with that operation was about 80%. Regarding your question about changing the treatment protocol, we do not have enough patients in one particular diagnostic group to be able to randomize them to different arms of treatment. I would hesitate to speculate about changing management based on the cost factors at this point. Like you, I do not define cure lightly. However, I think we can say with reasonable certainty that children with ventricular septal defects and coarctations can be cured in infancy and can become productive components of our society. That is the context in which I have used the word cure. You asked me to speculate about the indirect costs and how these are going to be handled. I wish I knew the answer to that. I am fairly confident, though, that if we as physicians are not involvedin the decision-making process for how those costs are to be handled, other people will make those decisions for us. This paper is basically a way of trying to get the discussion started. I hope others will take this information and criticize it highly, as I am sure some people will do after the information is made available. Personally, I prefer not to speculate. In terms of the three patients who had costs greater than $400,000: The outlook for such patients is not very good. Our patients were ventilator-bound in the hospital for more than 3 months-usually their entire lifetime. We had hoped to find a certain subset of patients that we could define early, so that we could determine that proceeding with operation is not worthwhile. We were unable to find that subset early, but I hope that failure at this point does not prevent us from trying in the future.